June 6, 2014
Chimps Understand Game Theory Better Than Humans In New Study
April Flowers for redOrbit.com - Your Universe Online
As humans, we tend to take great pride in our position at the top of the food chain. A new study from CalTech, however, shows that, when it comes to competition, we might not rank as high as our cousins, the chimps.
David Levine, professor of Economics at UCLA, defines game theory as the theory of social situations because it focuses on how groups of people interact with each other. There are two varieties of game theory, cooperative and non-cooperative.
Colin Camerer, CalTech's Robert Kirby Professor of Behavioral Economics, led the chimp / human study in game theory.
The research team devised a simple game they called the Inspection Game in which two players (either two chimps or two humans) are set up back to back, each facing their own computer screen. Each pair started the game by pressing a red circle on the monitor. Then they selected one of two blue boxes, one on the right side and one on the left. After both players had chosen, the computer showed them their opponent's choice. In each pair of players, one individual took on the role of the matcher by trying to choose the same as their opponent, while the other took on the role of the mismatcher by trying to choose the opposite of their opponent. The game consists of 200 iterations, with the winner gaining a small reward — a chunk of apple for the chimps or a small coin for the humans. To win repeatedly, a player has to accurately predict what their opponent will do next, learning to anticipate their strategy.
Despite being simple, the game simulates a common situation for both chimps and humans. For example, says Peter Bossaerts, visiting associate in finance at CalTech, think of an employee who only wants to work when her boss is watching. When the boss is not watching, she prefers to play video games. To hide her video game playing, the employee must learn to anticipate her boss's patterns of behavior. The employer, on the other hand, needs to be unpredictable in order to keep employees on their toes.
In addition to modeling such everyday situations, the Inspection Game provides methods for quantifying behavioral choices. "The nice thing about the game theory used in this study is that it allows you to boil down all of these situations to their strategic essence," explains Caltech graduate student Rahul Bhui.
If both opponents are playing strategically, there is a limit to how many times one can win. A majority of game theorists agree that the limit is best described by the Nash equilibrium, named for mathematician John Forbes Nash Jr., winner of the 1994 Nobel Memorial Prize in Economic Sciences, whose life and career provided the inspiration for the Academy Award–winning 2001 film A Beautiful Mind. Investopedia defines the Nash equilibrium as "a concept of game theory where the optimal outcome of a game is one where no player has an incentive to deviate from his or her chosen strategy after considering an opponent's choice."
The study was split into two experiments. In the first, six common chimpanzees (Pan troglodytes) and 16 Japanese students played the game (always against a member of their own species) at the Kyoto Primate facility. When the results of the two groups were compared, the researchers were surprised. Based on previous experience with human players, the humans this time played as expected. They slowly learned to anticipate their opponent, with reasonable results. They did not play optimally, however, and ended up somewhat shy of the Nash equilibrium.
On the other hand, the chimpanzees' performance was much better. The chimps learned the game more rapidly than their human counterparts, and nearly attained the predictions of the Nash theorem for optimal play. The chimps continued their impressive performance even as the researchers introduced changes into the game. First, the researchers had the chimps switch roles. Then the payoffs were changed. The matchers received a bigger reward for matching on one side of the screen instead of the other, which changed the Nash equilibrium for the game. The chimps changed their strategy, upholding their performance, right along with it.
The second experiment involved 12 adult men in Bossou, Guinea. In pairs again, the men were asked to face each other and given a bottle cap instead of a computer screen. Each man was to place their bottle cap either top up or top down. As with the first experiment, the players assumed the role of either matcher or mismatcher. The stakes were higher in Bossou, however. The winner received what amounted to approximately a full day's pay. Even with raised stakes, the men did not match the performance of the Kyoto chimps. The men in Bossou landed as far off the Nash equilibrium as the Japanese students who only received a handful of one yen coins.
There is more than one possible explanation for the enhanced performance of the chimpanzees. The chimps involved in the study have had extensive training at this kind of task, as well as more experience with the equipment used at the Primate facility than the humans did. The chimps in Kyoto were also related to each other. They played in mother-child pairs, which might have given them intimate knowledge, because of their long acquaintance, of the choices their opponent would likely make.
The researchers do not think either scenario explains the performance. The Japanese students might not have been familiar with the exact equipment used in Kyoto, but they were definitely acquainted with touch screens and video games. The men in Bossou were well acquainted with each other prior to the experiment, yet they performed no better than the students.
In previous Kyoto experiments, chimpanzees have been shown to have excellent short-term memory, which might be a factor in their superior performance. For example, in one game that pitted chimps against humans, the players were asked to watch a sequence of numbers flash briefly across a computer touch screen, then tap the squares in the sequence corresponding to the numbers they were shown. Humans find this task much more daunting than chimpanzees do.
The research team has suggested other reasons for chimpanzee superiority, however, at the Inspection Game. Two possible explanations exist: the roles of competition and cooperation in chimpanzee versus human societies; and the differential evolution of human and chimpanzee brains since our evolutionary paths split between 4 and 5 million years ago.
In the last 50 years, the debate over how "naturally" cooperative or competitive humans are has raged, and it is far from settled today. It is clear, however, that wherever humans rest on the cooperative / competitive spectrum, common chimpanzees are more competitive with each other than we are. Chimpanzees continually create and recreate a strong status and dominance heirarchy. The bonobo, Pan paniscus, is more highly cooperative than Pan troglodytes, but has not been studied as extensively. Humans, on the other hand, are highly prosocial and cooperative. This essential difference can be seen in chimp and human social development.
"While young chimpanzees hone their competitive skills with constant practice, playing hide-and-seek and wrestling," says Camerer, "their human counterparts shift at a young age from competition to cooperation using our special skill at language."
The researchers believe that language might play a key role in the results of the Inspection Game. During the game, the humans were not allowed to speak with one another, despite language being "key to human strategic interaction," according to Martin.
The second theory for the superior performance of the chimpanzees is called the "cognitive tradeoff hypothesis," and language plays a critical role here, as well. Tetsuro Matsuzawa, who developed this theory, says that the brain growth and specialization that led to distinctly human cognitive capabilities, such as language and categorization, could be responsible for humans processing simpler competitive situations such as the Inspection Game more abstractly. This would make the human response less automatic than the chimp response.
The current theories are speculative, however the researchers predict that emerging technologies will make it possible to "map out the set of brain circuits that humans and chimps rely upon so we can discover whether or not human strategic choices go down a longer pathway or get diffused into different parts of the brain compared to chimps."